SDXL Turbovs.AuraFlow v0.3

Which one is better?

In our own test, AuraFlow v0.3 leads. Neither is in the arena. Neither has an API price.

Test images

Same task, the first image of each. The task with the biggest difference is preselected.

SDXL Turbo: Character and detail, first image from the fixed test run — SDXL Turbo3 out of 100

AuraFlow v0.3: Character and detail, first image from the fixed test run — AuraFlow v0.316 out of 100

Result for “Character and detail”:AuraFlow v0.3 ahead

Prompt given to both modelsCreate a square cinematic full-body portrait of a fictional bicycle courier waiting under a transparent umbrella at a rainy tram stop in Hamburg at blue hour. She wears a mustard raincoat, navy trousers, red sneakers, a silver helmet, and carries a teal messenger bag. Keep both hands visible, render the bicycle correctly, use realistic wet-street reflections, and include no readable brand names.

What the rating looks at: Anatomy, clothing, and spatial consistency.

“Ahead”: better in all three rating rounds. “Slightly ahead”: a higher value, but not in every round.

Test protocol of the images

Each test attempt uses the same archived prompt and one request. The first output of that attempt counts. Repeats after technical test failures are disclosed for the affected images. Files are normalized to 1,024 × 1,024 pixels without cropping. Provider resolution, randomness, quality tier and prompt processing can still differ. The images are a practical visual test, not a reconstruction of the arena configuration.

Character and detail

SDXL Turbo: manual, local PyTorch 2.13.0, Diffusers 0.36.0; stabilityai/sdxl-turbo@71153311d3dbb46851df1931d3ca6e939de83304; native 1024x1024, 1:1; one request, first output; steps=1, guidance=0.0, seed=660064698563452219; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native tokenizer truncation: tokenizer 80 to 77 tokens, tokenizer_2 80 to 77 tokens; archived input unchanged, generated Nov 2, 2026, 6:23 PM. Source

AuraFlow v0.3: manual, local PyTorch 2.13.0, Diffusers 0.36.0; fal/AuraFlow-v0.3@2cd8588f04c886002be4571697d84654a50e3af3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=8761677709905735111; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:42 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 985a75b38642a9f0e5e0c0e0c67f1b7f67780d42bd2168fa24e9a7a7a94b0da6

SDXL Turbo image SHA 256: 9ee6443f4690649e93bdca408e5e8d317b312244636a3af47a02fe14eaf6c20c

AuraFlow v0.3 image SHA 256: 9d0f14ca671e82f18947b9a0d7d7587e372cb4f515ce028434fac4efbd937155

Infographic

SDXL Turbo: manual, local PyTorch 2.13.0, Diffusers 0.36.0; stabilityai/sdxl-turbo@71153311d3dbb46851df1931d3ca6e939de83304; native 1024x1024, 1:1; one request, first output; steps=1, guidance=0.0, seed=6060807684445796734; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:23 PM. Source

AuraFlow v0.3: manual, local PyTorch 2.13.0, Diffusers 0.36.0; fal/AuraFlow-v0.3@2cd8588f04c886002be4571697d84654a50e3af3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=3793893866514060497; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:47 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: ed65cb4e6a51e2336f8b149ad4cbb16e020f07d527ad94362e719d2dcf91d94f

SDXL Turbo image SHA 256: 20395d858252aa131f235baa1f514039503a52dfec958c8f5b299db708bf4090

AuraFlow v0.3 image SHA 256: d2a4ecc1c165bc861ca60133d9ff95dc9d5b1d5320763be92c8168ff8c3b8166

Product photography

SDXL Turbo: manual, local PyTorch 2.13.0, Diffusers 0.36.0; stabilityai/sdxl-turbo@71153311d3dbb46851df1931d3ca6e939de83304; native 1024x1024, 1:1; one request, first output; steps=1, guidance=0.0, seed=4654331835179858246; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE; native tokenizer truncation: tokenizer 80 to 77 tokens, tokenizer_2 80 to 77 tokens; archived input unchanged, generated Nov 2, 2026, 6:22 PM. Source

AuraFlow v0.3: manual, local PyTorch 2.13.0, Diffusers 0.36.0; fal/AuraFlow-v0.3@2cd8588f04c886002be4571697d84654a50e3af3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=4682738952817248216; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:37 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: c67fd71e91304458a2feef5c91efa8677260fab0c0099a29a2c52cb579f5b2f3

SDXL Turbo image SHA 256: 75e5b85283110815e51a0470c937c3e42407244b50d918a3c0c54e10998a4b4c

AuraFlow v0.3 image SHA 256: 6889065a427cc29bc1c0a53b0ad0662eb24f4be885148b0ebb8f6d945bbd4749

Typography and layout

SDXL Turbo: manual, local PyTorch 2.13.0, Diffusers 0.36.0; stabilityai/sdxl-turbo@71153311d3dbb46851df1931d3ca6e939de83304; native 1024x1024, 1:1; one request, first output; steps=1, guidance=0.0, seed=1511362670078321145; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:21 PM. Source

AuraFlow v0.3: manual, local PyTorch 2.13.0, Diffusers 0.36.0; fal/AuraFlow-v0.3@2cd8588f04c886002be4571697d84654a50e3af3; native 1024x1024, 1:1; one request, first output; steps=50, guidance=3.5, seed=7437917803059710684; mps for text encoder and denoising; cpu for native VAE decode; float16 for text encoder and transformer; float32 for VAE, generated Nov 2, 2026, 6:31 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 2e012c3d14ee5dd02203a94f394ca80a5038cdebd6032b25a8b8a221f2255419

SDXL Turbo image SHA 256: 963088c72f75c98a95fd71be5f8d6e14fdf17061ae984f1109522a39be24ac76

AuraFlow v0.3 image SHA 256: b6c221541210743564f6921b5e0bbc227c4d2246f6410cb0c1779c7c58d9460d

The numbers at a glance

Arena ratingfrom votes by users · higher is better

n/a

SDXL Turbo

not in the arena

n/a

AuraFlow v0.3

not in the arena

GraduallyBenchour own rating, 0 to 100 · higher is better

SDXL Turbo

Rank 105 of 108

AuraFlow v0.3

Rank 102 of 108

Price per imagedollars through the API · lower is better

n/a

SDXL Turbo

no API price

n/a

AuraFlow v0.3

no API price

Ranks among the 21 image models with an arena rating and the 108 with a GraduallyBench run; prices compared with 28 models with an API price.

SDXL Turbo

Subscription: n/a No plan in Germany names exactly this model.
Via API: No API price known. Access: local.
License: Open weights.

AuraFlow v0.3

Subscription: n/a No plan in Germany names exactly this model.
Via API: No API price known. Access: local.
License: Open weights.

Subscriptions only count plans that name exactly this model. Subscription prices include VAT, API prices are in dollars.

GraduallyBench from Nov 3, 2026 · Subscription database from Sep 29, 2026 (38 days old)

All data side by side

The better value is bold. For ratings only when the gap is clear.

Kennzahl	SDXL Turbo	AuraFlow v0.3
Ratings
Arena ratingfrom votes by users, higher is better	n/a	n/a
Arena rank	n/a	n/a
Arena votesmore votes, a safer rating	n/a	n/a
GraduallyBenchour own rating, 0 to 100	4	12
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition	3	11
Product photographyMaterials, reflections, and fine details	3	15
Character and detailAnatomy, clothing, and spatial consistency	3	16
InfographicNumbers, labels, and visual organization	8	7
API price
Per imagelower is better	n/a	n/a
100 images	n/a	n/a
Profile
Released	n/a	n/a
Weightsopen means: runs locally	open	open
Lifecycle	Older open fast checkpoint; no active hosted API ID claimed in the reviewed model list	no notice
Access	local	local
Parameters	not published	7 billion
Largest output	not consistently published	1,536 × 1,536 pixels
Plans in Germanynaming exactly this model	n/a	n/a

API prices for one image at 1,024 × 1,024 pixels according to Artificial Analysis.

Sources: SDXL Turbo model source, status source (verified Nov 1, 2026) · AuraFlow v0.3 model source (verified Nov 1, 2026)

Arena ratings by category

From blind votes on LMArena. “Ahead” means the lead is clear.

No shared arena score in the stored snapshot

The stored benchmark database contains no measurement for SDXL Turbo and AuraFlow v0.3 from the same arena and category. Older model versions are not used as substitutes.

More comparisons

Compare SDXL Turbo and AuraFlow v0.3 with the models people compare them with most often.

Compare another matchup

First image modelSecond image model